Virtualized Cloud Data Center Networks: Issues in Resource Management. by Linjiun Tsai & Wanjiun Liao

Author:Linjiun Tsai & Wanjiun Liao
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

The fault tolerance mechanisms can be much more resource-efficient if only one or few failures may occur at any point in time. Multiple services, even of different types, are allowed to share one or more single resource unit as their backup. An example is shown in Fig. 3.7, where three services of different types share one backup resource unit. Such simple but effective backup sharing mechanisms help raise resource utilization, no matter how complex the topologies requested by services. Even after reallocation (discussed in the next section), it is not required to find new backups for those reallocated services as long as they stay on the same axes. In data centers that are much more prone to failure, services are also allowed to be backed with multiple backup resource units to improve survivability, and those backups can still be shared among services or just dedicated. The ratio of these two types of backups may be determined according to the levels of fault tolerance requested by services or provisioned by data center operators.

Fig. 3.7An example of efficiently reserving backup resources for fault tolerance

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.